TF2-Wide&Deep-subclass

import matplotlib as mpl
import matplotlib.pyplot as plt
%matplotlib inline
import numpy as np
import sklearn
import pandas as pd
import os
import sys
import tensorflow as tf
from tensorflow import keras

#Get Data
from sklearn.datasets import fetch_california_housing
housing = fetch_california_housing()
data = housing.data
target = housing.target
print("data.shape = ",housing.data.shape)
print("target.shape = ",housing.target.shape)


#split train_test data
from sklearn.model_selection import train_test_split

x_train_all , x_test ,y_train_all , y_test = train_test_split(
    data,target,random_state=7,test_size=0.25
)
print("x_train_all.shape = ",x_train_all.shape)
x_train , x_valid , y_train,y_valid = train_test_split(
    x_train_all,y_train_all,random_state = 11,test_size=0.25
)
print("x_train.shape = ",x_train.shape)

data.shape =  (20640, 8)
target.shape =  (20640,)
x_train_all.shape =  (15480, 8)
x_train.shape =  (11610, 8)

#Normarlization
from sklearn.preprocessing import StandardScaler
scalar = StandardScaler()
x_train_scaled = scalar.fit_transform(x_train)
x_valid_scaled = scalar.transform(x_valid)
x_test_valid = scalar.transform(x_test)

# Built Model

# 子类法
class WideDeepModel(keras.models.Model):    #和pytorch的模型构建方法类似
    def __init__(self):
        super(WideDeepModel,self).__init__()
        """定义模型的层次"""
        self.hidden1_layer = keras.layers.Dense(30,activation="relu")
        self.hidden2_layer = keras.layers.Dense(30,activation="relu")
        self.output_layer = keras.layers.Dense(1)
        
    def call(self,input):
        """完成模型的正向计算"""
        hidden1 = self.hidden1_layer(input)
        hidden2 = self.hidden2_layer(hidden1)
        concat  = keras.layers.concatenate([input,hidden2])
        output = self.output_layer(concat)
        return output

model = WideDeepModel()
model.build(input_shape=(None,8))#None 表示样本数,8表示特征数



#查看模型
model.summary()

#编译模型
model.compile(loss="mean_squared_error",optimizer="sgd") 

Model: "wide_deep_model_2"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_6 (Dense)              multiple                  270       
_________________________________________________________________
dense_7 (Dense)              multiple                  930       
_________________________________________________________________
dense_8 (Dense)              multiple                  39        
=================================================================
Total params: 1,239
Trainable params: 1,239
Non-trainable params: 0
_________________________________________________________________

history = model.fit(
    x_train_scaled,
    y_train,
    validation_data=(x_valid_scaled,y_valid),
    epochs=10,
)

Train on 11610 samples, validate on 3870 samples
Epoch 1/10
11610/11610 [==============================] - 1s 88us/sample - loss: 2.0550 - val_loss: 0.8995
Epoch 2/10
11610/11610 [==============================] - 1s 70us/sample - loss: 0.7359 - val_loss: 0.7515
Epoch 3/10
11610/11610 [==============================] - 1s 67us/sample - loss: 0.6680 - val_loss: 0.7048
Epoch 4/10
11610/11610 [==============================] - 1s 67us/sample - loss: 0.6313 - val_loss: 0.6697
Epoch 5/10
11610/11610 [==============================] - 1s 68us/sample - loss: 0.6018 - val_loss: 0.6402
Epoch 6/10
11610/11610 [==============================] - 1s 68us/sample - loss: 0.5788 - val_loss: 0.6162
Epoch 7/10
11610/11610 [==============================] - 1s 71us/sample - loss: 0.5592 - val_loss: 0.5960
Epoch 8/10
11610/11610 [==============================] - 1s 70us/sample - loss: 0.5424 - val_loss: 0.5790
Epoch 9/10
11610/11610 [==============================] - 1s 69us/sample - loss: 0.5285 - val_loss: 0.5645
Epoch 10/10
11610/11610 [==============================] - 1s 69us/sample - loss: 0.5156 - val_loss: 0.5502

def plot_learning_curves(history):
    pd.DataFrame(history.history).plot(figsize=(10,6))
    plt.grid(True)
    plt.gca().set_ylim(0,1)
    plt.show()
plot_learning_curves(history)

model.evaluate(x_test_valid,y_test)

5160/5160 [==============================] - 0s 36us/sample - loss: 0.5292





0.5292306062787078